[Clang] Accept and forward `-fconvergent-functions` in the driver

Currently the `-fconvergent-functions` option is primarily used by GPU
toolchains to enforce convergent operations in line with the semantics.
This option previously was only supported via `-Xclang` and would show
up as unused if passed to the driver. This patch allows the driver to
forward it. This is mostly useful for users wishing to target GPU
toolchains directly via `--target=` without an offloading runtime.

Reviewed By: JonChesterfield, MaskRay

Differential Revision: https://reviews.llvm.org/D149019

diff --git a/clang/include/clang/Driver/Options.td b/clang/include/clang/Driver/Options.td
index 2abec84271356473c70c5efbd92cafc7b47b16b3..3b07244b98e82b9b950c2b97bd965f649b83de8a 100644 (file)
--- a/clang/include/clang/Driver/Options.td
+++ b/clang/include/clang/Driver/Options.td
@@ -967,8 +967,10 @@ def cxx_isystem : JoinedOrSeparate<["-"], "cxx-isystem">, Group<clang_i_Group>,
   MetaVarName<"<directory>">;
 def c : Flag<["-"], "c">, Flags<[NoXarchOption, FlangOption]>, Group<Action_Group>,
   HelpText<"Only run preprocess, compile, and assemble steps">;
-def fconvergent_functions : Flag<["-"], "fconvergent-functions">, Group<f_Group>, Flags<[CC1Option]>,
-  HelpText<"Assume functions may be convergent">;
+defm convergent_functions : BoolFOption<"convergent-functions",
+  LangOpts<"ConvergentFunctions">, DefaultFalse,
+  NegFlag<SetFalse, [], "Assume all functions may be convergent.">,
+  PosFlag<SetTrue, [CC1Option]>>;
 
 def gpu_use_aux_triple_only : Flag<["--"], "gpu-use-aux-triple-only">,
   InternalDriverOpt, HelpText<"Prepare '-aux-triple' only without populating "

diff --git a/clang/lib/Driver/ToolChains/Clang.cpp b/clang/lib/Driver/ToolChains/Clang.cpp
index edea4acfc3abddc753c1f14435b932de5a9de951..f7544c8f593c8d86194ce7f1443b7c8e6c786a76 100644 (file)
--- a/clang/lib/Driver/ToolChains/Clang.cpp
+++ b/clang/lib/Driver/ToolChains/Clang.cpp
@@ -5700,6 +5700,8 @@ void Clang::ConstructJob(Compilation &C, const JobAction &JA,
                     options::OPT_fno_unique_internal_linkage_names);
   Args.addOptInFlag(CmdArgs, options::OPT_funique_basic_block_section_names,
                     options::OPT_fno_unique_basic_block_section_names);
+  Args.addOptInFlag(CmdArgs, options::OPT_fconvergent_functions,
+                    options::OPT_fno_convergent_functions);
 
   if (Arg *A = Args.getLastArg(options::OPT_fsplit_machine_functions,
                                options::OPT_fno_split_machine_functions)) {

diff --git a/clang/lib/Frontend/CompilerInvocation.cpp b/clang/lib/Frontend/CompilerInvocation.cpp
index cb29049d5a3dd8088746475d0d04c7fd64967922..34e0e8d53e50ebb95ea70fb3aea0d30b74c8c60c 100644 (file)
--- a/clang/lib/Frontend/CompilerInvocation.cpp
+++ b/clang/lib/Frontend/CompilerInvocation.cpp
@@ -3731,9 +3731,9 @@ bool CompilerInvocation::ParseLangArgs(LangOptions &Opts, ArgList &Args,
   Opts.Blocks = Args.hasArg(OPT_fblocks) || (Opts.OpenCL
     && Opts.OpenCLVersion == 200);
 
-  Opts.ConvergentFunctions = Opts.OpenCL || (Opts.CUDA && Opts.CUDAIsDevice) ||
-                             Opts.SYCLIsDevice ||
-                             Args.hasArg(OPT_fconvergent_functions);
+  Opts.ConvergentFunctions = Args.hasArg(OPT_fconvergent_functions) ||
+                             Opts.OpenCL || (Opts.CUDA && Opts.CUDAIsDevice) ||
+                             Opts.SYCLIsDevice;
 
   Opts.NoBuiltin = Args.hasArg(OPT_fno_builtin) || Opts.Freestanding;
   if (!Opts.NoBuiltin)

diff --git a/clang/test/Driver/amdgpu-toolchain.c b/clang/test/Driver/amdgpu-toolchain.c
index 3477d50527c66659fea47ddd95dbe8e5cc95921b..b8b6667333d819f5a773edb2d42d2daa08c28f60 100644 (file)
--- a/clang/test/Driver/amdgpu-toolchain.c
+++ b/clang/test/Driver/amdgpu-toolchain.c
@@ -11,6 +11,6 @@
 // DWARF_VER: "-dwarf-version=5"
 
 // RUN: %clang -### --target=amdgcn-amd-amdhsa -mcpu=gfx906 -nogpulib \
-// RUN:   -flto %s 2>&1 | FileCheck -check-prefix=LTO %s
-// LTO: clang{{.*}} "-flto=full"
+// RUN:   -flto -fconvergent-functions %s 2>&1 | FileCheck -check-prefix=LTO %s
+// LTO: clang{{.*}} "-flto=full"{{.*}}"-fconvergent-functions"
 // LTO: ld.lld{{.*}}-plugin-opt=mcpu=gfx906